Skip to content

Latest commit

 

History

History
20 lines (11 loc) · 1.67 KB

README.md

File metadata and controls

20 lines (11 loc) · 1.67 KB
Raw

Replication of Hinton & Nowlan (1987) "How Learning Can Guide Evolution"

This is a repository with the code and article submitted to the ReScience journal.

The reference for the original article (APA style):
Hinton, G. E., & Nowlan, S. J. (1987). How learning can guide evolution. Complex systems, 1(3), 495-502.

The structure of the repository is the following. Folder article contains a brief description of my replication of the evolutionary simulations from the original article (see article/stojic2017.pdf).

Folder code contains main script simulation.R, written in R that runs the simulation, produces the summary data in CSV format that is saved in data folder. Script utils.R loads required packages and defines additional useful functions. Data in data folder is used by genFigure2.R in the code folder to produce the figures used in the replication article. Folder notebook does not contain anything.

How to run the simulations yourself?

The code has been developed on Linux operating system. To execute the code you will need to install R and couple of R packages. More details on operating system and R is available in code/README.md.

Note:

  1. Simulations take some time - on a 16 core computer it takes about 8h. For this reason, the data is already provided in data directory.
  2. If you execute code/simulation.R it will automatically try to install needed versions of the packages and potentially overwrite your version if you have some of them installed. Code might very well work with your version but I cannot guarantee it.